‘The General Purpose Agency Is Doomed’

Introduction:

This week, we’re joined by special guest Alan Pentz, who recently stepped back from his government-contracting business to start the Owner Institute, which draws on lessons he learned the hard way to help business owners scale their businesses. In his new role, Alan has immersed himself in the world of generative AI, and he’s come to some intriguing conclusions, one of which is that AI will eliminate most B-to-B agencies—marketing agencies, public relations agencies, professional services firms. Why is that? Because, Alan says, businesses will no longer be willing to pay agencies retainers of $5,000 or $10,000 a month once they realize they can get similar or even superior work from an AI chatbot. “In general,” Alan says, “most technology waves end up with a few big winners, and most people are just roadkill.” To explore the theory that agencies are likely to be roadkill, we invited Jaci Russo, owner of a marketing agency, and Sarah Segal, owner of a public relations agency, to have a conversation with Alan. Spoiler alert: there were no tears, no threats, and no insults.

— Loren Feldman

Guests:

Alan Pentz is founder of both Corner Alliance and the Owner Institute.

Jaci Russo is CEO of BrandRusso.

Sarah Segal is CEO of Segal Communications.

Producer:

Jess Thoubboron is founder of Blank Word.

Full Episode Transcript:

Loren Feldman:
Welcome Jaci, Sarah, and our special guest, Alan Pentz, who is founder of a couple of businesses we’re going to talk about. Glad to have you here, Alan.

Alan Pentz:
Oh, I’m glad to be the new guy on the pod. Thanks for having me, Loren.

Loren Feldman:
One of those businesses, Corner Alliance, you built to, I believe you’ve written, $35 million in annual revenue?

Alan Pentz:
Yeah, that was my first business, and we’re in government contracting, so that’s been a little challenging this year.

Loren Feldman:
I can imagine.

Alan Pentz:
Yeah, I ran that company for about 17 years, until I stepped down formally on January 1 as CEO.

Loren Feldman:
And when you stepped back from that business, you started another one called the Owner Institute, which offers guidance to small business owners. From what I’ve seen, it draws largely on the lessons you learned building Corner Alliance. You write a very savvy weekly blog post that I frequently highlight in the Morning Report, and I believe you’re working on building a generative AI tool that you are designing to guide owners through the process of building a business. Do I have that right?

Alan Pentz:
That’s right. That’s right. So the tool is called OwnerRx.com—like Owner Prescription. And when I launched it, I hadn’t really intended to build a company, but, of course, I’m an entrepreneur, so that’s the first thing I did. I got a hammer, and that’s a nail, right? And so I tried to do some coaching for small businesses. I got really frustrated with a couple of things. One, just quickly, the cost to make it worth my time. To take dedicated time on that, it’s very difficult to charge enough to hit the small business market that I want to serve. I could do that with groups, but trying to schedule five entrepreneurs is like herding cats with ADHD, right?

Loren Feldman:
Tell me about it. [Laughter]

Alan Pentz:
You’ve only got three on the podcast. You can’t get five, right?

Loren Feldman:
But I need a panel of 12 to get those three.

Alan Pentz:
That’s right. That’s right. Exactly. And then a lot of the planning methodologies out there, I felt like, are very process-based. So, you’ve got EOS, Scaling Up—they’re all great. But they actually don’t tell you how to do anything. And that’s what I was really seeking throughout my career, joining Entrepreneurs’ Organization and YPO and going to entrepreneur education events, hunting and pecking all over the place to get these lessons. So I really wanted to figure out a way to put them all in one place and then take advantage of the new technology out there. So I’m basically trying to build a strategic business coach that’s AI-powered in an app form. So that’s what I’m launching later this summer.

Loren Feldman:
That’s very exciting. I’m sure we will talk more about that, but we’re here today because I attended a webinar you offered recently to help owners create their own AI GPTs. During the webinar, kind of as an aside, I think, you noted that you believe AI is going to force small agencies—marketing agencies, PR firms, and the like—to alter their business models pretty dramatically. Can you explain what you have in mind with that?

Alan Pentz:
Yeah, I mean, I quoted Sam Altman, the CEO of OpenAI from, I don’t know, last month sometime, saying that he believes 95 percent of what ad agencies do today, marketing agencies, will be done by the AIs. And I think he’s saying, basically, the general LLMs—although that might be a little much. I’m in it all day long, and it’s really driven by the newer models, which cost a little more. They’re $200 a month instead of $20. I get the sense that they make me 10 to 100 times more productive.

And when you start working in them every day, and you start realizing what you can do, you start thinking. I’m reflecting on the small businesses I’m working with, and I’m like: Oh my God, you are going to be able to reach a thousand times more customers, but you’re going to have to do it at a tenth or a hundredth of the price, because people can do a lot of this on their own now. And they’ll be able to go to the AI and make it happen.

And if you’re charging $3,000 to $5,000 to $10,000 a month, it’s worth it for someone to go and get the AI to do it for them. You charge them 50 bucks a month, they probably aren’t going to bother. They’ll just have you do it. But with that 50 bucks, now I can serve a thousand clients, right? And so I think that’s going to be the tradeoff we see. And we’re just going to see the vast stack of work that agencies do just get eliminated. I mean, go listen to Mark Zuckerberg talking about what Meta is going to do. You don’t even have ad copy. You just say, “I sell this.” And they go tell you exactly, they basically do it all for you and deliver your customers. That’s what’s going to come.

Loren Feldman:
So Jaci and Sarah, you both own the kinds of agencies Alan is talking about. Any questions?

Jaci Russo:
I agree 100 percent.

Sarah Segal:
I agree to a point. I believe that it’s going to increase productivity 100 percent and that we will be able to deliver as an agency more results and content and project deliverables because of AI. But I don’t think that it’s going to eliminate our value to small businesses, because—at least for my business—most of the time that we’re brought on board with a small- to mid-size business, is when the internal team goes, “You know what? We’ve been trying to do this ourselves. It keeps getting put on the back burner. We need somebody else to take this off our plate and somebody who knows what they’re doing.” It’s one thing to say, “We can give you the tool to do marketing or PR.” It’s another thing to know what you’re doing. And you still need that skill-set. And so that is the value of an agency, regardless of what AI does for you. And that’s my take on that.

Alan Pentz:
You know, PR might be slightly different. I wrote that from a marketing agency standpoint. But it’s my sense that it’s not necessarily the client themselves. It’s that there will become competitors who are able to do 95 percent of what you do for 50 bucks a month. That’s what I think is gonna happen. You’re gonna have a competitor that’s embedded all those—basically what I’m doing with small business coaching, right? Take all the knowledge I have, put it into either a wrapper around the AI, or eventually, they’ll probably build their own models. It’s going to be the competitor that comes in and just does that 95 percent, not necessarily the client.

Sarah Segal:
Yeah, but still, what I do is relationship-based. We’ve had this conversation on the podcast and in real life many times, where AI is great, but it can’t have relationships. And when you’re talking to journalists, influencers, and doing that part of the job, I don’t know that AI is going to ever be able to replace that. Because the reason why I get coverage, or the reason why we can tell a story about our client or work with an influencer is because we’ve established a relationship with them. I know a lot of reporters and influencers now that are really irritated by AI and that they’re getting these crappy processed pitches that they can see that are AI-driven 100 miles away. And it’s off-putting to them.

Alan Pentz:
I’ll let Jaci respond, too, but I do have a response to that.

Jaci Russo:
Why don’t you all keep going, and I’ll jump in with my own paddle on the Ping Pong table after? [Laughter]

Alan Pentz:
Yeah, I mean, I think that’s very true today. One thing I would do is think through how your entire industry is going to change. So influencers and news are already going to change completely. So I think to get to the conclusion of what the PR agency will do, we’ve actually got to look at what’s going to happen to journalism, news gathering, and influencing. And I think that that whole thing is getting changed. The whole idea of this open web is going away. So people are going to be doing these things at scale in and for the actual LLMs, for the AI companies. So I think we’re moving to a different kind of internet.

Sarah Segal:
Oh, 100 percent.

Alan Pentz:
And at that point, I think people are going to be able to do things tailored at scale that they’re not able to do today. So for instance, I would say the journalist of the future is probably writing for an AI and all they’re doing is taking all these submissions and figuring out through whatever algorithm what’s going to perform best to get them the goal they’re trying to get with communicating the message in the AI. So I actually think the tools and everything, the way we do it around it, will completely transform.

Sarah Segal:
Yeah, I mean, journalism is always transforming. It has gone from this one report or one story a day kind of thing, to where it’s more crowd-sourced information at this point. I have two teenaged children who never read the news, never look at anything. They get their information through TikTok. And then most kids these days, they get it through TikTok, and then they’ll verify it through a third party—another source, hopefully. But with that, you still need a reporter, a journalist, on the scene, or other people who are contributing to that story on the scene.

I think I agree with you that it’s gonna transform, and then, as an agency, I need to make sure that I’m incorporating AI into what I’m doing. And I am doing that, and I’m doing it now to create efficiencies. And we’re keeping up with what’s happening. But I think old-school PR professionals are going to be in trouble if they don’t adopt AI, 100 percent. But I don’t think that it’s going to impact businesses that kind of see it coming. When I started, we were pitching people via fax. [Laughter] We had to figure out how to do things differently over time. So there’s always a transformation. 50 years from now, it’s going to be a whole different playing field as well. So I think that we’re naturally inclined to keep up with that.

Alan Pentz:
Well, that’s good, and I would agree. So I said 95 percent. I do think there’s going to be 5 percent that will probably be relationship, knowledge based. And I think the ability to charge for that will actually go up. So I do think there’s going to be a judgment relationship. The other thing I’ve talked about is building communities, gathering data about your customer set and whoever.

So, Sarah, for example, I would say: Could I figure out a way to build in, a way to do whatever percent, 85 percent of what you do, into an application, which I can now basically write myself by prompting so that it takes the client auto through most of the stuff? There’s a few places where you’re using relationships, but by scaling that offering to more people, you get more data more specific about your customer set, what they’re doing, what they care about, what works. That data is yours. That’s not going into the LLM. And you create community data, and then you have experience and judgment, and that’s where you’re going to get that 5-percent value that the LLMs won’t get. And then you can charge a lot for that. So I don’t think it’s total doom and gloom story. I just think that most of it’s going to get commodified.

Loren Feldman:
I’m dying to hear—

Sarah Segal:
—what Jaci says.

Loren Feldman:
Jaci, why do you agree with Alan? And what are you doing about it with your agency, if you agree with him?

Jaci Russo:
Well, I’ll actually start at a different point and then handle those two. As you have heard me mention a time or two, Loren, I am angry that my industry doesn’t take itself as seriously as every other professional service industry does and that we’ve never put in place education and licensing and certifications the way that contractors and architects and accountants and hair stylists and nail technicians and, my gosh, every single one of them. This is how that problem gets solved, because this is going to be the thing that eliminates, I’ll go with 95 percent of the and—I’m using air quotes here—”agencies” that exist right now. Because so many of them are doing entry-level work that has already been replaced by AI.

So, I’m glad. I think it’s going to clear out, in some ways, a lot of the entities that don’t have the strategic mindset, the value partnership, truly bringing something to the table. They’re now going to have to find new ways to elevate and I’m not in charge of them, so I’m not worried about it, but that’s going to change the landscape. So on that hand, check, I’m happy. I agree. I think it’s a good thing.

And the answer to your other questions: I think that it’s going to be a problem for a minute, because it’s going to have more clutter, because we can all create so much more content than ever before, so much faster, that there’s going to be a lot more stuff. There’s already a lot of stuff now. There’s going to be even more stuff. And I see the benefit of that, because the really good stuff, the human-inspired, still had a human-involved, still feels kind of personal and real, that’s gonna rise to the top. Because the good stuff always rises to the top. So, I think that’s good.

Last but not least, what am I doing about it? A couple of things. You and I have talked for probably two years now about how we’ve incorporated AI here. We’ve ramped that up even more so this year, and I agree with Alan’s point about the more you pay, the better the quality is. And so we’re seeing that on our side as well. We are not just using it to make ourselves more efficient, which I think is always the first step. We are now starting to use it to help our clients be better at things outside of just marketing. So it’s going to end up probably being a different entity, but I think that it is the way we provide value. Because value is no longer, “Let me just write a blog post for you or make a social media post for you or do a design for you.” Those things can all be done in the AI. So, how do we elevate our value and our strategic partnership? That’s where we get to stay in business.

Loren Feldman:
So Jaci, you don’t think it’s going to have a huge impact on the way you run your business, other than infusing AI into everything you do, but not necessarily changing your business model? Is that fair?

Jaci Russo:
Well, I think the business model continues to evolve. Just me, personally, you want to talk about how this agency has changed in 25 years? On February 1, 2001, this agency was a freelance media buyer who was nine months pregnant and needed to be able to take a maternity leave. And her former employer, as of January 31, 2001, told her that she could have a maternity leave if she scheduled the delivery on a Friday and was back on a Monday. [Laughter] So that’s how this agency started. So sure, we’ve evolved. Our freelance media buyer in this story had five clients. They were restaurants, furniture stores, and car dealers, and we bought TV, radio, newspaper, and billboards. I’m trying to think of the last time I did that. [Laughter] You know, it’s been a minute. Those aren’t our clients, and that’s not the work we do.

And so we evolved into a full-service agency, because I am married to a graphic designer. And so Michael left his agency where he was the art director and came to this agency. So now it’s an agency, because it’s two people. And then we evolved again, because we saw social media coming like a ton of bricks and knew we weren’t just gonna buy TV and radio and newspaper anymore. And so we evolved again.

Well, now we’re just tactical, and we didn’t like that. So then it became: Let’s do what we’re really good at: strategic brand planning in the B-to-B space. So it’s professional service companies, manufacturers, it’s industrial. Because not enough people were paying attention to them, and we had a real strength there. And I’m looking at the year-over-year gains of our clients. There’s an ROI that keeps us around, even though we’re on a 30-day-out clause with every one of them. There’s no contracts, and our average lifespan is six and a half years with our clients.

Alan Pentz:
Wow. Go work with Jaci.

Jaci Russo:
Right? So here’s another opportunity to evolve and grow, and I’m excited about it. I like coming to this building every day, and I like working with these people, but every single person who works in this building will tell you what they are hired for and what they do today—not the same thing. Because we evolve and grow.

Alan Pentz:
I think the one thing that’s coming that we haven’t really talked about is—and I’m gonna use the horrible catch phrase buzzword—the “agentic web.”

Loren Feldman:
Alan, I’m going to interrupt you for a second. Some of our listeners might appreciate it if you define what you mean by an agent.

Alan Pentz:
So right now, if you go into ChatGPT, there’s a thing called operator. You push on it and you say, “Hey, go open this website and try to book my plane ticket.” And, like, it’s a mess. It doesn’t really work. But what we’re moving toward is an open standard that will allow United Airlines and American Airlines to build a website that’s just for agents, so that they can autonomously come in on your instructions and book your plane ticket for you. So that’s what we’re moving to.

Loren Feldman:
Maybe this reflects my bias as an old school journalist, but here’s my question, I guess, first for Alan, and then for Jaci and Sarah as well. I hear what you’re saying, Alan. You’ve acknowledged that relationships are, in fact, important, and there’s room for judgment there. I would go even a little bit further and say: Well, we know that AI is great with structuring information, and it’s great with analyzing information. We really haven’t seen that creative spark out of it yet. Maybe it’s coming, I don’t know, and maybe you’re saying that that would be represented in that 5 percent of the agency business that goes on more or less the way it is now. And I guess I just want to push back a little bit and say: Isn’t that more than 5 percent of the industry?

Alan Pentz:
So I guess, to be clear, it’s not 5 percent of the value. It’s 5 percent of the current activities that happen, work effort. So it can be 50 percent of the value. I could argue it might be even more. The other thing I would push back is, I’ve seen the spark of creativity, no question.

Loren Feldman:
Tell me about that. How so?

Alan Pentz:
So I don’t think it’s like 100 percent the AI alone. It’s the AI working with me. But the AI is bringing things to the table that have transformed my business multiple times. And I would say this is definitely more working with the most advanced models and the more expensive subscriptions. I think, if you’re in professional services, and you’re not paying 200 bucks a month to get o3 and o4-mini, come on. It’s so valuable. And I’ve had creation sessions with it that have blown my mind and created entirely new ways and things I’m going to do with my company.

Loren Feldman:
We need an example. Alan.

Alan Pentz:
So for example, I started talking about, “Hey, can I build this small-business app to give my lessons and help people do it in small chunks every day?” It says, “Yes.” I’m like, “What else could I do with it?” Well, it came up with the idea of, “Why don’t you make it into a platform that other people can bring their ways of doing things, and you become the platform and they sell in their own user journeys and ways of doing things?” So right there, it’s like: Oh my god, it just came up with a major pivot to this business that could blow it into a whole different category. So it’s not like I wasn’t involved in it, but it brought it to the table, and I didn’t think of it.

The other example I’d give you is like I did in the webinar. I did that board of titans, Loren. I think AI is very good at taking personas. So I set up a persona for Jeff Bezos to come in and talk about customer-centricity, Elon Musk to talk about first principles, blah, blah, blah, a bunch of other titans who are known. Every strategic idea I put through that, it’s blown me away. Every time, it upgrades what I’m doing. So I think the creativity is there. It doesn’t mean that humans don’t have a part of that. I think it requires a human-AI interaction. But it isn’t just a dumb agent that does research for you.

Loren Feldman:
Jaci or Sarah, any thoughts?

Sarah Segal:
I mean, I don’t think it’s a dumb agent. I just think that there’s a human element to creativity that is not replaceable. And I think no matter how much technology exists, you still need that human element to make it better, to make it interesting, to connect with people. Because that’s what people connect with when they’re looking at an ad or a post or—

Alan Pentz:
Can I interrupt and ask a question?

Sarah Segal:
Yeah.

Alan Pentz:
So this is what I heard Mark Zuckerberg talking about the other day. What if you—this is a marketing example. So, sorry. It’s not your industry.

Sarah Segal:
Go for it.

Alan Pentz:
But he said: Look, you can come in and say, I target B-to-B companies for this, for public relations, whatever it is. And you bring in no copy. You just have a customer you want to reach. And to be honest, they’re now getting better at finding the customers that will buy your services than you are. And they just create a thousand different versions of every possible permutation of how they would get someone to buy your thing. And then they see what works, and they just put more money behind that. And then they deliver you more than any creativity you alone could ever come up with an ad and put it out. So they’re just taking care of that by doing it a thousand times.

Sarah Segal:
So in the PR industry, when we talk about pushing out a news story, there are people who pray and spray, is what it’s called, where they just blast out pitches. And it sounds like AI will help customize that to the audiences a little bit more. But I just worry about the recipient. It’s less about whether or not it’s going to help me scale my execution, but it’s how people on the other end are receiving it. And are people ready? Are people seeing it? Do they like it? Do they want it? I’m getting AI email messages and phone calls at this point, and I delete and I hang up because it’s not a real person. And I don’t want to not talk to a real person.

Alan Pentz:
Yeah, I think the answer to that is—first of all, what’s happening today, I’m not going to say, “Hey, this is perfect,” right? And the answer to AI is more AI. So you’re going to have AIs on the other end that are helping you sift through things—the five things that you like as human. So, what I see is the AIs get better at talking to each other, and they get better at surfacing the things you want. Now, again, I’m not going to say there’s no human creativity involved here, like my examples of collaborating. I look at it as collaborating with the AI as a strategic partner. It’s like having my own McKinsey next to me.

So it doesn’t take me out of the loop, but it just radically transforms the content. It allows me to make leaps I never could have made. It radically transforms what I can do. I can now program apps that I can serve to people in a couple of days. So it’s just different. It’s going to change all our business models, and we need to start thinking through: We’ve captured what we do that’s unique. We are embedding it into technology that can scale it way past our current client set. That allows us to collect data. And how do we build community? And so that’s what I think the professional service owner needs to recognize—not that no human will ever be involved, but that these are going to allow people to come in and take your customers for 50 bucks a month, if you don’t do it first. That’s sort of the warning I come with.

Loren Feldman:
Jaci, have you seen the spark of creativity in AI? And do you see the potential that Alan’s talking about?

Jaci Russo:
I would say yes. An example of that would be, Loren, you may recall that last year on the Brand State U side—Alan, I run two companies. I run the strategic BtoB agency BrandRUSSO, and then an offshoot of that that’s been in place for about five or six years now called Brand State U that is online, primarily, marketing training is how it started, for people who can’t afford an agency. See, there’s a pivot. We got bigger, we got better, our prices got higher. We had to start saying no to people who wanted to hire us, but they couldn’t afford us. So now I can send them somewhere else to get a different kind of help, so they can learn to do it for themselves with a little bit of guidance. So that’s kind of taking care of both ends of the spectrum. It’s worked really well.

So last year, the Brand State U team did something we called the GrowthX conference, and it was two days with first-stage businesses that know enough to have started up, but don’t know enough to scale. Huge, huge positive feedback. So my brainstorming partner, ChatGPT and I, spent some time talking about why that went well, reviewing the survey results from the participants, talking about what this looks like, how it can grow, what it should grow into. And it is now GrowthX Academy, which will launch in September. We’ve got two states—one state firmly committed, one state I think they’ll be committing in the next week or two—to do a pair of six-month cohorts that is going to be a blend of two-day retreat, six months of weekly sessions with subject matter experts covering the seven pillars of business.

I maybe would have eventually thought of that on my own, but the creative brainstorming of me saying, “This is who we’re doing it for. What do they need?” “Well, these are the problems.” “Okay, well, how do I help solve those problems?” “Okay, what about this?” “Well, what would those people object to?” “Well, these would be their objections.” “Well, then how to overcome those objections?” You know, going through that process that we all do when we’re working with our AIs. All of a sudden, I’ve got a fully fleshed out, built-out curriculum that got run through subject matter experts who tweaked it and modified it for where their expertise is. I made some wild guesses in the finance area, because that’s not my strong suit, and was pleasantly course-corrected by the CPA who’s teaching the classes. And I can’t wait to attend his stuff, because there’s some things I obviously need to learn.

So it’s not just like the AI is teaching us, which I do still have some pauses on, because I’ve seen it with Brand State U—the, “Here’s the information. Go teach yourself.” We still need connectivity. We still need community. We still need cohorts. We still need peers. And so this is a blend of that. I don’t know that I would have come up with it without that brainstorming partner.

Alan Pentz:
I think, Jaci, that’s awesome. And what I see is the evolution of that. So what you did with Brand State U, I could see that becoming the basis of creating agents. So instead of people going to classes anymore, you just have an agent—it’s like a program that walks them through how to create all these things for themselves and then operates it for them. So they tailor it to their needs based on your curriculum. But it’s not like going to a class and then you go off and do stuff. It’s like you’re just building it right there, and then you have a fully operating business at the end.

Jaci Russo:
And that’s what we’re doing through this. Each of these sessions is incorporating AI, teaching the prompts. So for example, in the finance section, they’re going to learn it, because they run these companies. They’re the founders and the CEOs. They need to understand it. But here are the prompts for you to take your financials, plug it in with these prompts to understand so that you’re not just learning it today; you’re learning how to learn it six months, a year, three years.

Alan Pentz:
So here’s what I want, Jaci, I would love to see that business become like the wrapper around all that. So they don’t just get prompts. You have a platform that helps them build all their stuff. It’s stored there, and they operate out of there. And then now, all of a sudden, you’re the SaaS provider to all these people, helping them adopt AI and power their business. And so instead of a course, you’re now building a SaaS application that everyone customizes to themselves, and it’s all powered by AI. That’s where I think we’re going in the future.

Sarah Segal:
Can I share a prediction?

Alan Pentz:
Yeah.

Sarah Segal:
So my feeling about a lot of marketing- and/or service-based industries is that you get what you pay for. And if it’s too good to be true, it probably is. And my feeling about AI becoming so accessible to everyone in the world is that there’s going to be a lot of crap out there for a long time where people are like, “Oh yeah, for 50 bucks a month, we’re going to get you in The New York Times.” Or, “For 50 bucks a month, we will increase your followers on your social channels and drive X amount of click-throughs to your website.” And there’s going to be a lot to wade through.

And we’ve seen this time and time again, where the agencies will promise amazing results, and people will buy into it and then be disappointed by it and disillusioned by the industry. So I think that there needs to be some caution taken by the industry, and marketing in general, in that people are going to sign on for some of these things, and they’re not going to be well-managed by a human like Jaci. They’re going to be disappointed in the outcome. And so I’m anticipating a lot of crap for a while before we weed out the quick-result applications.

Alan Pentz:
I wouldn’t disagree that there’s going to be a lot of crap for a while. That I fully embrace. I just do think that a lot of the things we’re doing will not require humans to be a part of it. Yes, there’s a human overall in your campaign and as a strategist and partnering with the AI. But a lot of the interaction that you have junior people doing, or even mid-level managers, will just be agent-to-agent. And then at that point, you get what you pay for, because there’s no human involved. It’s just agent-to-agent, so it’s scalable. And I think we’ve seen this. Software has done this in industry after industry, and I think it’s just going to happen here.

So I think, the key I see is the person who’s able to serve more and more clients and collect data that they know that the LLMs don’t have, that person is going to be more valuable. They have a viewpoint, they have experience, and they have data, and that’s where you’re going to pay. You’re going to pay double, triple, quadruple for that. That’s what I urge people to start doing now, is setting it up so you get the data.

Loren Feldman:
Alan, let me ask you this. I think we’re all kind of agreeing here that there’s a future for the best in the industry, but you described kind of two alternatives. One is being among those who are able to kind of continue on, mostly the way they are now, and the majority, who have to find a different business model, probably not able to get the big retainers, having to charge less but have a lot more clients. For somebody who’s listening to this, who owns an agency, and who isn’t confident that they’re going to be among the best, for all those others, that sounds like a race to the bottom to me. And most of those other agencies are just going to disappear. Is that how you see it?

Alan Pentz:
Yeah, I mean, welcome to the social media-ization of everything, right? So, I mean, that’s what we’ve seen over and over with tech, with some huge winners and a bunch of people who go out of business. I do think, in this case, what do you do? It’s pretty typical stuff. So if you go on OwnerRX and go through the business model exercise, it’ll tell you you’ve gotta niche out. You gotta figure out: Can I get a better niche? Can I do something very specific that I can then make more?

And again, think about it this way: If I can learn how to do something, I can now scale it to millions of people within days. We’re not quite there, but we’ll be there soon. And so then I start collecting data about that thing and that niche, and I can use that to create more value. That’s where the niche smaller players will survive. But yeah, I think you’re right in general. Most technology waves end up with a few big winners, and most people are just roadkill.

Loren Feldman:
So if one of those companies comes to you, looking for your help, either as a one-on-one consultant or through your AI GPT, what are they going to be told?

Alan Pentz:
I mean, I think you have got to figure out your niches, whether that’s a customer base or a thing you do. You’ve gotta then embed that into software, and you’ve gotta create a mechanism for you to build your own data layer and start figuring out how I scale this to more people. The general purpose agency is doomed.

Sarah Segal:
I actually think that there’s a positive side to this that needs to be articulated, that the AI is going to help us do so many things that we shouldn’t be spending our time on. Like, when I first started out in television, I would do an interview with an individual and have to sit there and log that interview to figure out the sound bites, the time codes, all that kind of stuff before I put the story together. Now that’s instantaneous, so reporters can spend their time on other things.

I think the same goes for agencies. We’re not all of a sudden going to become obsolete. We’re just all of a sudden going to find all this time where, “Oh, you know what? We can do this, and we can do this, and we can expand on this project that we’ve been wanting to do.” We want to do more live, in-person, events. It’s going to help us spend our time more wisely if we do it correctly.

So I see AI as not a doom-and-gloom, like it’s going to take our jobs, but more like it’s going to take away the the the stuff that we shouldn’t be wasting our time on, and the brainless stuff that a two-year-old can do—even like taking notes, right? We all take meeting notes when you’re on an agency call. I’m sure Jaci has automated this already, and that they have their Otter or what have you taking notes and writing summaries of it. We do that too—to a limit, though, because there are nuances in what we discuss with our clients that aren’t picked up well on any of the note trackers. But yeah, why would we have someone do that all the time when they can do something else?

Alan Pentz:
I just came up with an analogy while you were talking, Sarah. So I think this is going to do to agencies what social media did to journalism. So there will be a New York Times, there will be a Wall Street Journal, and not much else. And then we’re going to have the Substacks and the bloggers and an explosion of these specialists who are building little communities and things around them, so it will help democratize that part of it. Newspapers were based on classified ads and 976- numbers, and they just got wiped out. That’s what’s coming. It’s that level of destruction. I agree with all the stuff about, “Hey, it’s going to take away the stuff we don’t want to do.” I think it’s going to fundamentally change the business model.

Sarah Segal:
So it’s like what the internet did to Encyclopedia Britannica.

Jaci Russo:
And the Yellow Pages.

Alan Pentz:
Maybe, maybe. I would say, I would say, a little bit. There’s still a New York Times. There’s still a Washington Post. There’s still a Wall Street Journal. So maybe it won’t be extinction for everything, but it’s really going to take a few people who are high agency, who do the stuff I’m talking about and reinforce their win. And I think it’s going to hollow out most of the rest. But you are going to be able to build a business around your unique thing and your take and your community and the data you collect, but it’s going to have to be highly differentiated and highly niche.

Jaci Russo:
I think about lamplighters a lot, probably more than I should, but really, there used to be—

Alan Pentz:
Have you been watching Mary Poppins?

Jaci Russo:
I have not, although I might. But there used to be people, and that was their job. And it was a necessary and important job. And electricity came along, and we didn’t need people to go light the lamps anymore. I would imagine, sitting around some dinner tables, there were some people who were very concerned about their future, because that is how they put food on the table for their family. However, I think the world is in a better place for having electricity, and I would imagine that those lamp lighters became chimney sweeps or electricians or plumbers or something else that utilized their skills.

I don’t think this is an extinction-level event for the people who do the jobs. I think the jobs change. Every time I read the doom-and-gloom articles about unemployment, I’m like, “Really?” Did unemployment skyrocket when lamplighters went out of business? No. When Ford developed the assembly line? No. People find new ways to use their skills. And the great thing is, the tool that’s eliminating your ability or need to do this one job is the same tool that can help you learn a new skill to do a new job. So we should benefit from it, basically.

Alan Pentz:
So well said. Totally agree.

Loren Feldman:
Alan, I think we’ve been talking primarily about how this affects smaller agencies that tend to have smaller B-to-B clients. What do you think it means for big corporations? Is Coca Cola going to have agency work that it handles the same way it always has? Or are big companies going to make this change, too?

Alan Pentz:
Yeah, I mean, I think they are. It’s going to start from the bottom up. I don’t think it’s going to start at the top down. So it’s going to eat away. Again, it’ll get there someday. I think they are going to be the ones that pay the big, big money for the decisional people who have that experience and knowledge. So there’s clearly going to be a place where someone’s making millions a year, just advising on what the AIs are doing and looking at the data and saying, “No, I don’t think this, I think that”—and taste-making kind of things. So that last 5 percent is 5 percent of the work right now, but it’s going to become 50 percent of the dollars, maybe, or something like that. But, yeah, if they see that things are done just as well by an AI, why are they going to pay people to do it?

Sarah Segal:
I have a question that kind of stems off that. I think about it as malls and mom-and-pop shops, where the malls are the big, enormous agencies that we see. And then you have the mom-and-pop shops, which is more along my size. Is AI going to change how people hire agencies? Will they hire smaller agencies, knowing that those agencies can do the same thing that the larger agency does, because they have the same tools and AI but they’re going to get more attention—as opposed to hiring the big agency, where they are just a name and a number? I don’t know.

Alan Pentz:
Well, what I see in the future, if we get this layer that’s sort of the new internet for agents, I kind of see, you’re a person at a company. You’ve got to do a campaign. And so you go to some place with your AI, and it has all the agents that say: We get these kind of outcomes at this cost. And you’re kind of going to an app store almost, or it’s automated somehow for you. And you’re selecting, so it probably won’t even matter what size person or group of people built that. It’ll be the outcomes it gets for this very, very, very specific niche. And then you select it and pay for it based on that. I think it’s all agent-based at some point.

And then there’s this layer, where someone’s super famous for this idea, and they have this persona, and they have this knowledge, and then corporations contract with them, where agents are doing 99 percent of work, and they’re just kind of a person architecting the campaign, or bringing a new perspective. That’s how I see it kind of playing out.

Sarah Segal:
Interesting.

Loren Feldman:
We’re just about out of time. Alan, I’m curious, you started a business 20 years ago, and now you’re starting one here in 2025 at the dawning of the AI age. You’ve already indicated to us that you’ve been surprised at the things you can do now that obviously you couldn’t even dream of 20 years ago. Are there other things like that? Are there other things you are able to do today that might surprise other business owners who haven’t started a business recently?

Alan Pentz:
Yeah, Loren, I was telling you before, my new website, I went into a program called Replit, where it’s one of these coding agents. It’s a little easier to use than some of the more advanced ones.

Loren Feldman:
Where you don’t have to be a coder to code?

Alan Pentz:
That’s right. So I went in and said, “Hey, I want a modern-looking, SaaS-like website. I gave it my logo and designer fonts and our colors, and it built me a website in an hour. And it’s up at OwnerRx.com if you want to look at it. And it was amazing. And then I started building an app today to help record leadership team meetings and take you through a method, sort of like an EOS kind of method or something, of an L 10 Meeting. It helps you go through the meeting, record everything, and then distributes the action items and all that. They call it vibe-coding now in Replit, and within a week or so, I’ll be able to release it and have people use it. So it’s incredible. You can have an idea, and within, I would say a few days, you can do basic applications and websites. I would say within a year or two, you’re going to go from idea to being able to serve an app to millions of people in hours.

Loren Feldman:
Jaci, when Alan said that, did you look at OwnerRx?

Jaci Russo:
I haven’t yet, because I always think the thing is going to fuss at me if I’m not paying attention to your podcast platform. It gives low-engagement warnings. I wrote it down.

Alan Pentz:
Stay focused. Stay focused.

Sarah Segal:
I’m looking at it. So you created this using prompts?

Alan Pentz:
Yeah, so five minutes before the meeting, I had it so that it was like, “Sign up for the product while I’m building my waitlist right now.” So I was like, “Oh, change the button to ‘Join the Waitlist,’ and use this landing-page address.” And I did that. Two minutes later, it redeployed, and I came on the podcast.

Sarah Segal:
But you have testimonials.

Alan Pentz:
Those are fake. [Laughter]

Sarah Segal:
Yeah!

Alan Pentz:
I don’t have a product out yet. But when, when I get the testimonials—I’m doing an alpha group pretty soon. I just signed up some people, and as they go, I’m going to capture that stuff with their picture, and I just go back in and say, “Hey, replace the testimonial with what this person said in this picture.” And it goes and does it in five minutes. It’s incredible. So, yeah, caveat, not everything on the site—I put it up today, so there are a couple things I still need to work on.

But, yeah, like I said, this application actually has a database. It has fields you go through. It can plug into your Otter, or your Fathom, or whatever notetaker you have and then auto go through the notes of your leadership team meeting, put it in there, pull out the action items, assign them to people. And then over time, using Gemini API, I can have it look at trends and blindspots and other things in your leadership team meetings. And, oh, by the way, you’re going to go into your next quarterly meeting, you just go back in and say, “Hey, I’m doing my quarterly meeting,” and I’ll hit the AI and it’ll say, “Here are all the trends and things you talked about. Here’s what you accomplished. Here’s what you didn’t accomplish. And here’s our recommendation for what you should talk about in your quarterly meeting.” I mean, I can do that in days.

Sarah Segal:
What did you build this through again?

Alan Pentz:
This is Replit. You know, it can get frustrating. Anything you have to stop and start, it’ll choke on certain things. So you gotta be somebody who’s willing to work with it. But you can do pretty basic applications in those, and then you can move them over to things like Cursor, which are what real coders are using. But even if you just want to mock something up and it doesn’t work perfectly, you can mock it up and hand it to a developer and cut half the development process.

Loren Feldman:
Jaci, what do you think?

Jaci Russo:
I’m fascinated.

Alan Pentz:
It’s a pretty good website, right?

Jaci Russo:
Yeah!

Loren Feldman:
That’s it, Jaci?

Jaci Russo:
Well, Loren, I’m paying attention, so I’m not able to really dig into it. I signed up for his list. I started going through his type form to get the feedback thing to see how that works. But I mean, I see the similarities between what I’m doing with Brand State U and what he’s doing with OwnerRx, and how, by being a couple of years later, he’s a decade smarter, because there’s more technology now to do some things way better than we did five years ago.

Alan Pentz:
Yeah, well, I would tell you, Jaci, I couldn’t have done this four months ago. So it is changing that quickly. It’s a different game than it was three months ago.

Loren Feldman:
All right, we are out of time. My thanks to Jaci Russo, Sarah Segal, and especially Alan Pentz. Alan, you gave us a lot to think about here. Thank you.

Alan Pentz:
I appreciate it. I like being the new guy. It was really fun. Hopefully, I didn’t make everybody want to jump off a bridge.

Jaci Russo:
No!

Sarah Segal:
A small bridge. [Laughter]

Loren Feldman:
Thanks, everybody.